Instructions to use nurlanboranbay/tonton with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ultralytics
How to use nurlanboranbay/tonton with ultralytics:
from huggingface_hub import hf_hub_download from ultralytics import YOLO # pick the weights file from this repo's "Files and versions" tab weights = hf_hub_download("nurlanboranbay/tonton", "<weights>.pt") model = YOLO(weights) source = 'http://images.cocodataset.org/val2017/000000039769.jpg' model.predict(source=source, save=True) - Notebooks
- Google Colab
- Kaggle
tonton3 models
The weights used by the tonton3 cash-desk violation service (ForteBank Datathon 26). Code: https://github.com/NurlanBoranbay/tonton
| file | what | size |
|---|---|---|
tonton3/yolo26x-pose.pt |
people + 17 keypoints, tracking (Ultralytics YOLO26) | 126 MB |
tonton3/yolo26x.pt |
COCO-80 objects: laptop, cup, bottle, handbag... (Ultralytics YOLO26) | 119 MB |
tonton3/custom_y26s_all_stamp.pt |
our 11-class detector (YOLO26-s, 960 px), trained on all 42 videos: money_stack, collection_bag, counting_machine, safe_open, safe_closed, key_in_lock, key_out_of_lock, door_open, door_closed, ID-identification, stamp | 100 MB |
sh tools/download_weights.sh # from the GitHub repo; puts the files in weights/
- Downloads last month
- 453